基于集成学习的航空器着陆跑道占用时间预测
Aircraft runway occupancy time prediction based on ensemble learning
陈亚青 1陈九龙2
作者信息
- 1. 中国民用航空飞行学院空中交通管理学院,广汉 618307
- 2. 中国民用航空飞行学院民航飞行技术与飞行安全科研基地,广汉 618307
- 折叠
摘要
为准确预测航空器着陆跑道占用时间,提出了一种基于集成学习的航空器着陆跑道占用时间预测模型.首先,根据航空器机载QAR数据处理后得到的航空器影响因素,利用皮尔逊积矩相关系数进行相关性排序;其次构建了基于stacking集成学习策略的预测模型,实现了对着陆跑道占用时间的预测;最后通过模型评价指标对比了各预测模型预测精度.实验结果表明,提出的集成学习预测模型的准确性更高,可为机场实际运行效率提升提供理论依据.
Abstract
Arrival runway occupancy time prediction model based on ensemble learning is proposed to accurately predict the time of an aircraft occupying the runway during landing.Firstly,the influencing factors are obtained by processing the onboard QAR(Quick Access Recorder)data,and their correlation is ranked using the Pearson product-moment correlation coefficient.Secondly,a prediction model based on stacking ensemble learning strategy is constructed to predict the arrival runway occupancy time.Fi-nally,the prediction accuracy of different models is compared by calculating model evaluation metrics.Experimental results demon-strate that the proposed ensemble learning prediction model achieves higher accuracy and provides a theoretical basis for improving the operational efficiency of airports.
关键词
集成学习/跑道占用时间/预测模型/QARKey words
ensemble learning/runway occupancy time/prediction model/QAR引用本文复制引用
出版年
2024
NETL
NSTL
万方数据